Convergence analysis on time scales for HOBAM neural networks in the Stepanov-like weighted pseudo almost automorphic space

• Published in: Neural computing & applications 2021-04, Vol.33 (8), p.3567-3581
• Main Authors: Arbi, Adnène, Guo, Yingxin, Cao, Jinde
• Format: Article
• Language: English
• Subjects: Artificial Intelligence; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Convergence; Data Mining and Knowledge Discovery; Feasibility; Fixed points (mathematics); Image Processing and Computer Vision; Linear algebra; Mathematical models; Neural networks; Original Article; Probability and Statistics in Computer Science; Recurrent neural networks
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-020-05183-0

We start by presenting the concept of Stepanov-like weighted pseudo almost automorphic on time-space scales. Besides, we introduce a novel model of high-order BAM neural networks with mixed delays. To the best of our knowledge, this is the first time to study the convergence analysis for one system modeling the recurrent neural networks. Since it is a Δ -dynamic system on time-space scales, the results obtained in our work are new and attractive. The main difficulty in our theoretical work is the construction of Lyapunov–Krasovskii functional and the application of the Banach’s fixed-point theorem. By fabricating an appropriate Lyapunov–Krasovskii Functional (LKF), some new sufficient conditions are obtained in terms of linear algebraic equations to guarantee the convergence to the Stepanov-like WPAA on time-space scales solution for the labeled neural networks solutions. The obtained conditions are expressed in terms of algebraic equations whose feasibility can be checked easily by a simple calculus. Furthermore, we have collated our effort with foregoing one in the available literatures and showed that it is less conserved. Finally, two numerical examples show the feasibility of our theoretical outcomes.